What proportion of South Australian adult non-vegetarians hold similar beliefs to vegetarians?

Abstract

(Nutr Diet 2004;61:11-21)

Objective: The aim of this study was to determine the proportion of non-vegetarians with similar beliefs as vegetarians, and to examine their personal characteristics.

Design: Mail survey on meat and vegetarianism beliefs, personal values and information sources.

Subjects and setting: One thousand people were randomly selected from the South Australian population via the telephone directory. Approximately 600 questionnaires were obtained (71% of those able to be contacted participated). An additional non-random survey of 106 vegetarians and semi-vegetarians was conducted, totalling 707 participants.

Main outcome measures: (1) Identification of the proportion of non-vegetarian respondents with similar beliefs as vegetarians ('cognitive vegetarians'); (2) Examination of their personal characteristics and a comparison with vegetarians and the remaining non-vegetarians (e.g. meat consumption, interest in vegetarianism, personal values, use of and trust in information sources).

Statistical analyses: Categorical principal components analysis was performed and a plot of object points labelled by dietary group identified cognitive vegetarians. The characteristics of cognitive vegetarians, vegetarians and the remaining non-vegetarians were examined using cross-tabulation analyses.

Results: We found that 15% of non-vegetarians shared similar beliefs about meat and vegetarianism as vegetarians. Cognitive vegetarians formed a distinct group, e.g. they were less likely than the remaining non-vegetarians to eat red meat as frequently or identify as Anglo-Australian, but were more likely to be female and interested in vegetarianism.

Conclusion: A significant proportion of the population may be susceptible to the promotion of vegetarianism and, hence, plant-based diets. This has implications for health professionals and promoters of plant-based and vegetarian diets.

Key words: vegetarian diet, social psychology, meat, questionnaires, Australia

**********

Introduction

Vegetarianism is a dietary practice that incorporates a multitude of interrelated attitudes, beliefs and nutritional behaviours. A vegetarian diet excludes meat, fish and seafood. However, there are some people who identify as 'vegetarian' despite including some animal flesh (particularly fish and fowl) in their diet. In addition, some people who eat comparatively small amounts of meat or fish refer to themselves as 'semi-vegetarian' (i.e. a low meat consumer). Estimates of the prevalence of vegetarianism vary from country to country. Epidemiological data are missing, but public opinion polls and other surveys indicate that 2.5% of the USA population (1), 3.7% of Australians (2), and 5.0% of the British population are vegetarian (3). There is some evidence that the prevalence of vegetarianism may be increasing. For example, in the USA 1% of the population was reported as being vegetarian in 1997 (4), while 2.1% of the British population identified as vegetarian in 1984 (5).

Although the numbers of vegetarians are at present relatively low, it appears that many more people choose not to eat meat--or at least some kinds of meat--to some extent. There has been a decrease in the consumption of red meat in recent years (6,7) and a study in the USA found that 57% of the population consume a vegetarian meal sometimes, often or always when eating out (8).

Diets that emphasise greater consumption of plant foods are increasingly being recognised by researchers and government and health organisations as providing important health benefits (9-12). Diets that emphasise plant foods usually contain low levels of saturated fat, cholesterol and animal protein, and are high in folate, antioxidants, fibre, carotenoids and phytochemicals (9). A number of studies have shown that, compared to non-vegetarians, vegetarians have a lower mean body mass index and plasma total cholesterol concentration (10,13-15), and mortality from ischaemic heart disease is decreased by about a quarter (10).

Vegetarian diets can provide adequate amounts of essential amino acids, iron, vitamin B12, calcium, vitamin D and zinc at all stages of the life cycle, but they must be well planned (9). They should include sufficient quantities of a variety of fruits, vegetables, legumes, nuts, seeds and grains (16,17). In addition, low-fat dairy products and eggs are useful sources of nutrients such as calcium and protein. Poorly planned vegetarian diets may entail the risk of not meeting the needs for some nutrients, particularly iron, zinc, and vitamin B12 (18). Although mean serum ferritin levels are lower in vegetarians than non-vegetarians, clinical anaemia does not appear to be more common among vegetarians (18,19), at least in western countries (20).

While the proportion of people with established vegetarian practice is known, there appears to be little information about the individuals who consume omnivorous diets who hold similar beliefs about meat and vegetarian diets as vegetarians. The current and future practices of these individuals have implications for health status, food consumption patterns and the food industry.

Methods

Procedure

One thousand South Australians were selected at random by utilising the Marketing Pro software package (Desktop Marketing Systems Pty Ltd, Melbourne, Marketing Pro, April 1999) containing a comprehensive list of South Australian residences from the telephone directory, in the form of names, addresses and telephone numbers. A 12-page booklet was mailed to each person in the sample together with a cover letter and reply-paid envelope. Dillman's recommended survey methods formed the basis of the questionnaire format and administration (21). A number of follow-up reminders were mailed to the sample to ensure the highest possible response rate, as described elsewhere (22).

A non-random sample (n = 106) of vegetarians and semi-vegetarians was also included in the survey because there was only a small number of vegetarians (n = 9) present in the random sample. The sample was selected by advertising in vegetarian cafes and restaurants, health food stores, by handing out questionnaires at an environmental fair at the University of Adelaide, and by word-of-mouth publicity.

The study was conducted in accordance with the guidelines of the University of Adelaide Human Ethics Committee.

Questionnaire

The pilot was formulated through an iterative process that was generated through interviews with 15 vegetarians, from the literature (23-26), and from existing food choice questionnaires. The latter were an Australian survey on meat consumption and attitudes (27); the Institute of European Food Studies (IEFS) European Union survey on attitudes to food, health and nutrition (28); and New Zealand and Australian surveys of concerns about food (29).

The questionnaire contained several sections including:

1. Beliefs about meat (20 items) were measured, with some items being based on the Worsley et al. questionnaire (27). Respondents answered by indicating their level of agreement with each belief on a five-point Likert-type scale, ranging between 'strongly disagree' and 'strongly agree'. Some examples of the belief statements are: 'I love to eat red meat such as beef, veal or lamb'; 'Meat is necessary in the adult human diet'; 'Meat causes cancer'; and 'Meat production is cruel to animals'. Table 1 lists the complete set of belief items.

2. Perceived benefits of vegetarian diets (24 items) were assessed, parts of which were modelled on the IEFS survey (28). Respondents specified their level of agreement or disagreement with each benefit on a five-point scale. Some of the benefits were: 'Control my weight'; 'Eat more fruit and vegetables'; 'Help animal welfare/rights'; and 'Lower my chances of getting food poisoning' (see Table 2 for the full list of items).

3. Personal barriers to vegetarian diets (27 items) were measured, some of which were modelled on the IEFS survey (28). Respondents indicated their level of agreement or disagreement with each barrier on a five-point scale. Barriers included: 'There is not enough iron in vegetarian diets'; 'I like eating meat'; 'There is too limited a choice when I eat out'; and 'I lack the right cooking skills' (see Table 3 for a complete list of the items).

The remaining sections were about respondents' frequency of use of and trust in sources of information about food, nutrition and health; recall of promotion of meat and vegetarianism by these information sources; self-identified dietary status (vegetarian, semi-vegetarian, non-vegetarian), (e.g. 'Are you a vegetarian now? Please circle one number. 1. No. 2. Yes. 3. I normally think of myself as being a semi-vegetarian.'). Items about vegetarianism in general included number of vegetarian friends and family and likelihood of becoming vegetarian in the future; beliefs about nutrition; frequency of consumption of meat and other animal products. Demographic variables included sex, age, education, occupational status and income; food and health concerns (based on Worsley and Scott (29); and personal values (based on Schwartz's Values Invetory (26)). Values may be defined as lasting beliefs about desirable goals that function as guiding principles in life (26,31). Some examples of the values included in the questionnaire are 'Protecting the environment', 'A varied life', and 'Respect for tradition' (26). The food and health concerns included 'Clean handling of food in shops', 'The irradiation of foods', 'Driftnet fishing', 'Genetic modification of foods', and 'Poverty in Australia'.

Note that, as reported previously by the authors (30), the question on frequency of consumption of meat allowed an examination of the relationship between self-perceived vegetarian status and actual (reported) behaviour. Self-identified semi-vegetarians had lower meat consumption than non-vegetarians (e.g. 49% ate red meat rarely or never compared to 4% of non-vegetarians) and the majority (85%) of self-identified vegetarians never ate meat, fish or seafood. As reported in Figure 2, 97% of self-identified vegetarians never ate red meat, while 3% rarely ate it. With white meat, 95% of self-identified vegetarians never ate it and 5% rarely ate it; while 85% never ate fish or seafood, 9% rarely ate it, 5% ate it one to three times per month, and 1% ate it one to four times per week.

Data analysis

Categorical principal components analysis (CATPCA) was performed on the correspondents' responses to three sections of the questionnaire, namely beliefs about meat, barriers to vegetarian diets and benefits of vegetarian diets. CATPCA is a multivariate statistical technique that quantifies categorical variables and reduces data dimensions. Object principal normalisation was used to normalise the object scores and the variables, as this method is useful for identifying any differences or similarities between objects (i.e. between cases). A plot of object points labelled by dietary group (self-identified vegetarian, n = 103; self-identified semi-vegetarian, n = 55; self-identified non-vegetarian, n = 546; an additional three respondents did not complete the question on dietary status although from the items on frequency of meat consumption they could be classified as non-vegetarian) was obtained. We employed data provided by individuals as to whether or not they identified themselves as belonging to a particular dietary group, rather than that obtained on frequency of their meat and fish consumption. This was because individual beliefs may be closer to self-definitions than to the 'objective' definition of vegetarianism (25). In addition, the majority of self-identified vegetarians consumed no meat, fish or seafood and therefore would be classified as vegetarian by using the conventional definition. Cronbach's Alpha was used to measure the internal reliability of the dimensions of the plot, based on the average correlation between items. From this plot, the non-vegetarians who fell into the same range as the vegetarians (V) were identified (i.e. those non-vegetarians who hold similar beliefs about meat and vegetarian diets as vegetarians). This group of non-vegetarians was named cognitive vegetarians (CV), in contrast to those non-vegetarians who were outside the range of vegetarians: omnivores (O). Although semi-vegetarians were included in the plot, they were not included in the graphical display of the plot (Figure 1) or in further analyses. They were excluded for reasons of brevity and to ensure that cognitive vegetarians remain the main focus of this paper.

Cross-tabulation analyses (including Pearson chi-square test of statistical significance) were performed to examine the responses of CV compared with O and V for meat and animal product consumption, items about vegetarianism including interest in vegetarianism and number of vegetarian significant others, nutrition beliefs, use of and trust in food/health/nutrition information sources, concerns about the food supply, personal values and demographic items.

Analyses were conducted with SPSS for Windows statistical software (SPSS Inc, Chicago, SPSS for Windows, version 10.0.05 1999), using an alpha level of 0.05 for all statistical tests. The number of significance tests conducted in this study introduces the possibility of Type I statistical error (i.e. the probability of incorrectly rejecting the null hypothesis). The results report significance levels of 0.001, 0.01 and 0.05 so that the reader can judge which level is most appropriate to satisfy any concerns about Type I error.

Results

Response and respondent characteristics

About 71% of the randomly selected subjects who were able to be contacted answered the questionnaire (n = 603), with two questionnaires being unusable. Approximately 15% of the sample was unable to be contacted because their addresses had changed since the Marketing Pro data were collected or were incomplete, or they could not be contacted by telephone. With the 106 non-randomly selected vegetarians and semi-vegetarians included, the total number of usable questionnaires was 707.

The demographic characteristics of the non-randomly selected sample, the randomly selected sample, and the general South Australian population, as obtained from the Basic Community Profile of the 1996 Census of Population and Housing (32), have been reported elsewhere (30). Comparisons were made for sex, age, country of birth, employment status, and marital status, as these were the only items that allowed for direct comparison between the Basic Community Profile and the survey. In summary, the major biases in the random sample were over-representation of 45 to 64 year olds and married people and under-representation of 19 to 24 year olds compared with the census data. The non-random sample was younger than both the random and census groups. In addition, members of the non-random group were less likely to be in full-time employment or to be married or widowed/divorced.

Prevalence of cognitive vegetarians

Figure 1 displays the areas covered by the object points for the vegetarian group and the two non-vegetarian groups (cognitive vegetarians and omnivores). Two dimensions were identified, both with high internal reliability as measured by Cronbach's Alpha (0.98 and 0.88 for Dimension 1 and 2 respectively). Dimension 1 differentiates reasonably well between vegetarians and non-vegetarians.

As would be expected, the majority of self-identified semi-vegetarians (about 75%) were in the vegetarian range for the CATPCA of beliefs about meat, benefits of and barriers to vegetarianism.

Approximately 15% of non-vegetarians (n = 81, equivalent to 13.5% of the random population sample) fell within the range of vegetarians, and were therefore classified as cognitive vegetarians. Cognitive vegetarians formed a group distinct from the remaining non-vegetarians (n = 468) and from vegetarians (n = 103).

Characteristics of cognitive vegetarians

Cognitive vegetarians were less likely to eat red meat as frequently as O (P < 0.001), and more likely to eat it than V (P < 0.001). Figure 2 displays the differences in frequency of red meat consumption between V, CV and O. Cognitive vegetarians were over three times as likely as O to consume red meat rarely and over twice as likely to consume it only one to three times a month. However, the majority of CV (54%) and O (71%) ate red meat one to four times a week. Cognitive vegetarians were also less likely to consume dairy products as frequently as O (P < 0.01), although they were more likely to consume them than V were (P < 0.001). However, there were no statistically significant differences between CV and O for white meat, fish/seafood or egg consumption. Differences between CV and V were all significant (P < 0.001) for these food items, with V consuming them least often.

Cognitive vegetarians were more interested in vegetarianism than O (P < 0.001). They were twice as likely as O to be somewhat interested (54% versus 27%), and almost 10% were very interested compared to about 2% of O (Figure 3). Cognitive vegetarians had talked about vegetarianism more often than O (P < 0.001); had thought more in the past month about making changes towards consuming a vegetarian diet (P < 0.001); were more likely to think that during the next month they would make some changes towards becoming vegetarian (P < 0.001) and were also more likely to think that during the next six months they would make such changes (P < 0.001). Cognitive vegetarians were three times as likely as O to state that it was quite likely that they would make some changes during the next six months, and about 3% of CV stated that it was very likely, but no O did (Figure 4). However, the majority of both CV (73%) and O (87%) stated that they were not likely to make some changes. Also of note is that over 10% of CV were unsure whether or not they would make any changes towards vegetarianism. This may indicate that they are somewhat ambivalent about vegetarianism, unlike O who were more against the idea of becoming vegetarian. Cognitive vegetarians had more vegetarian friends than O (P < 0.01) but less than V (P < 0.001), as shown in Figure 5.

Compared with the remainder of the non-vegetarians (O), CV were more likely to search for information about healthy eating, with 74% of CV agreeing that they frequently look for such information and 58% of O similarly (Table 4). There was no significant difference between V and CV. Cognitive vegetarians were the least likely of the three groups to agree with 'I do not need to make any changes to the food I eat as it is already healthy enough', with V most likely to agree (Table 4). Differences between V and CV were statistically significant, and although differences between CV and O were not significant at the 0.05 level, they were close to significance (P = 0.058).

In general, despite CV falling inside the CATPCA plot's range for V, CV were closer to O than to V for use of information sources to find out about food, health or nutrition, trust in information sources, personal values, and this was particularly true for food concerns. However, there were exceptions, some of which are displayed in Table 4. (Note that in this table and in the text only a selection of the survey items is shown or discussed, in order to high-light the characteristics of the CV group in as succinct a way as possible. Full results may be obtained from the authors.) Cognitive vegetarians were usually intermediate between V and O, and therefore could be clearly differentiated from both groups.

There were differences between V, CV and O for use of and trust in sources of food/nutrition/health information (Table 4). In most cases, CV were intermediate between V and O, although CV used and trusted dietitians and used cooking magazines more than the other two groups. For example, 23% of O trusted vegetarian magazines compared with 44% of CV and 81% of V. There was no significant difference between V and CV for use of and trust in health food shops and staff. Doctors were highly trusted by over 80% of both CV and O, but only 34% of V. Advertising was trusted by 4% of V, and by 18% and 20% of CV and O respectively. Although the difference between CV and O was not significant at the 0.05 level, it neared significance (P = 0.056).

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

There were differences between the personal values of CV, V and O. 'Protecting the environment', 'A world of beauty' and being 'Helpful' were more important guiding principles for CV than for O, with 65%, 48% and 55% of CV respectively stating that they were very important (Table 4). An example of a difference between V and the non-vegetarians is 'Equality', where almost 90% of V stated that it is very important, compared to 65% of CV and 60% of O.

There was little difference between CV and O for concerns about food, with only two concerns being (statistically) significantly different. These were: (1) concern over 'Chemical additives in food', which CV were less concerned about than O, with V more concerned about than O; and (2) concern about 'The links between food and cancer', with there being no statistically significant difference between CV and V, who were significantly more concerned about such links than O (Table 4).

With regard to demographic variables, CV were more likely than O to be female (P < 0.05); about 69% of CV were female compared to 54% of O. Cognitive vegetarians were also more likely to be born outside of Australia (P < 0.01); approximately 37% of CV and 24% of O were born in another country. They were less likely to identify as an Anglo-Australian (55% of CV identified as Anglo-Australian compared to 69% of O; P < 0.05), and were less likely to be employed full-time (19% of CV and 32% of O were employed full-time; P < 0.01). There were no statistically significant age, education or income differences.

Discussion

We have found that 1.5% and 7.2% of the South Australian population identify as vegetarian and semi-vegetarian respectively (unpublished observations), although the 1995 National Nutrition Survey reported that a higher proportion (3.3%) of the South Australian population is vegetarian (33). With the addition of the 13.5% of the population who may be classified as cognitive vegetarians, as reported here, a substantial minority (22.2%) of Australians appear to have a positive attitude towards vegetarianism and have either already become vegetarian or a low meat consumer, or may be susceptible to the promotion of vegetarianism (e.g. if some of the social environmental barriers to vegetarianism were removed). To our knowledge, there are no comparable data available at present on attitudes towards vegetarianism among omnivores in similar countries such as the USA or UK. This would be an interesting area for future research and would enable comparisons to be made between different countries.

Those who are likely to be susceptible to the adoption of vegetarian practices are also likely to be people who already consume a greater proportion of plant foods than the general population. However, the present study did not obtain data on the frequency of consumption of plant foods, so further studies are required to verify this. It may be expected that the proportion of omnivores susceptible to the promotion of plant-based diets that contain some meat, such as that advocated by the World Cancer Research Fund (12), would be higher than that susceptible to the promotion of the stricter diet of vegetarianism. If future research confirms this, it should be very encouraging for health professionals who advocate plant-based diets. There may also be implications for the food industry.

Cognitive vegetarians search for information on healthy eating more than omnivores. This may indicate that they have found more information on vegetarianism and hence are more accepting of it than omnivores. It also suggests that if information on vegetarianism and health was more widely available, cognitive vegetarians may be more readily influenced to become vegetarian, particularly if such information specifically addressed their concerns with, and perceived benefits of, vegetarianism. The finding that about half of CV--more than the other two groups--believe that they need to improve the healthiness of their diet supports the notion that they are open to dietary change. As this was a cross-sectional study, it could not be determined whether any actual dietary changes were (or will be) made subsequent to the survey by the CV group. Therefore there is a need for long-term tracking of people's behaviours (i.e. longitudinal research).

It is logical that consumers that fall in the same range as vegetarians for beliefs about meat, and perceived benefits of and barriers to vegetarianism would have lower red meat consumption than those who do not. Red meat is usually the first animal product to be avoided by meat reducers, followed by white meat and then fish (23,34). Research has found that often people who avoid red meat only will have an increased consumption of white meat and fish (34). However, this study found no difference in frequency of white meat and fish/seafood consumption between cognitive vegetarians and omnivores. Dairy consumption was lower, perhaps because of a perception of high fat content and other negative health associations (35). This may be linked to the differing demographic characteristics of the cognitive vegetarians, such as the increased likelihood of being female or of non-Anglo descent.

Significant others often influence beliefs and behaviour, such as intention to eat beef (36,37). This may have been the case for the cognitive vegetarians, as they had significantly more vegetarian friends than did the omnivores. However, a longitudinal study would be required to determine if this was the case or, rather, if the cognitive vegetarians had gravitated towards people who were similar to themselves.

Vegetarians are generally more in favour of unorthodox sources of health and nutrition information than non-vegetarians (24,38). In this study, it was vegetarian magazines and health food shops, but not alternative health practitioners, where the more important differences between CV and O lay. Vegetarian magazines are, of course, focussed specifically on vegetarianism; perhaps health food shops are also perceived as being vegetarian oriented. In conjunction with their (moderate) use of and trust in unorthodox sources, however, CV also trust orthodox sources highly, particularly when compared to V. It seems that CV believe both types of sources have something to offer, although usually orthodox sources are used and trusted more than unorthodox sources.

Research has found that vegetarians tend to hold different values to omnivores. In particular, universal values such as equality and protecting the environment are more important among vegetarians (27,39). This study found that CV held some universal and other values to be more important than did O, although many other values, including equality, were similar to those of O. Again, the question is whether these differences resulted from the beliefs of CV about meat and vegetarianism, or whether the value differences partially influenced their beliefs.

The finding that links between cancer and food are of more concern to CV (and V) than O suggests one possibility for the differing beliefs and behaviours of CV to other meat eaters. Cognitive vegetarians may associate high meat and low fruit, vegetable and wholegrain consumption with a greater risk of developing cancer, as per the World Cancer Research Fund and American Institute for Cancer Research (12). This possibility needs to be explored in future research.

Turning to demography, the lower proportion of CV employed full-time compared to O may be partly explained by the higher proportion of women in this group, since women are less likely to be employed full-time than men (32). Generally, more women than men hold negative attitudes towards meat and are vegetarians or meat reducers (2,40,41). Therefore, it is not surprising that a larger proportion of CV were women compared to O. Meat, particularly red meat, is often associated with masculinity: meat is perceived as men's food in many cultures, while non-meat foods such as vegetables are perceived as women's food (42). This is certainly the case in Australia (e.g. the 'feed the man meat' advertising campaign of a few years ago). In addition, women are generally more health conscious than men, which may lead to greater appreciation of fruits and vegetables in particular (43). They tend to have more knowledge about the healthiness of different foods, prefer healthier foods, and tend to rate them as more pleasurable than men do (44).

Historically Australia has been a country of high meat consumption (6,45), particularly when compared to Asian and Mediterranean countries. Meat is a more central part of the diet than complex carbohydrates (46). This may be why CV were less likely to identify as Anglo-Australian or to be born in Australia.

Conclusion

It appears that a substantial proportion of Australian omnivores hold similar beliefs about meat and vegetarian diets as vegetarians. To our knowledge this appears to be the first study to establish this connection. Clearly, our study has implications for the health status of these individuals and for the food industry. It is essential that the former are sufficiently knowledgeable about the potential health benefits and risks of vegetarian diets, while the latter will need to respond to any potential or actual dietary shifts in the population. Compared to other non-vegetarians, cognitive vegetarians differ in their meat consumption behaviour, demographic characteristics, beliefs, values, information seeking behaviour, and level of trust in information sources. These characteristics may be important to consider if attempts are made to influence meat beliefs and consumption.

Table 1. Percent of self-identified vegetarians and non-vegetarians in
agreement with statements of beliefs about meat items

                                   Vegetarians   Non-vegetarians
Item(a)                            (n = 103)      (n = 546)        P
                                   Percent in agreement (% unsure)

I love to eat red meat such as
beef, veal or lamb.                    0 (1)          82 (6)        ***
I prefer to eat red meat more
than fruit or vegetables.              0 (1)          22 (13)       ***
I think meat is disgusting            71 (12)          3 (5)        ***
Nothing satisfies my appetite
like a thick, juicy steak.             0 (0)          30 (13)       ***
Meat such as beef and lamb is
unhealthy to eat.                     64 (13)          3 (9)        ***
Red meat such as beef or lamb is
fattening.                            46 (28)         11 (15)       ***
Meat is necessary in the adult
human diet.                            1 (3)          68 (16)       ***
Meat is important for the
health of babies and toddlers.         2 (12)         42 (40)       ***
Meat is necessary in children's
diets.                                 0 (11)         67 (25)       ***
Meat causes cancer.                   46 (37)          2 (35)       ***
Meat causes heart disease.            55 (28)          7 (35)       ***
Red meat is the best-absorbed
source of dietary iron.               17 (26)         52 (39)       ***
Non-vegetarians are healthier
than vegetarians.                      2 (4)          15 (46)       ***
Red meat is very expensive.           30 (57)         47 (12)       ***
When I eat out to celebrate
a social occasion, I usually
eat some kind of red meat.             0 (0)          44 (11)       ***
Meat is important for building
strength.                              5 (7)          62 (27)       ***
Meat is more a food that men are
likely to eat.                        35 (24)         52 (19)       **
Meat production is cruel to
animals.                              93 (4)          18 (26)       ***
Humans have no right to kill
animals for food.                     69 (16)          8 (13)       ***
Vegetarians are hippies or
weirdos.                               7 (4)           8 (13)       *

(a) Items used in categorical principal component analysis.
***P < 0.001; **P < 0.01; *P < 0.05

Table 2. Percent of self-identified vegetarians and non-vegetarians in
agreement with statements of perceived benefits of vegetarianism

                                  Vegetarians   Non-vegetarians
Item (a)                          (n = 103)       (n = 546)          P
                                  Percent in agreement (% unsure)

Control my weight                    56 (23)         38 (30)         ***
Prevent disease in general (e.g.
heart disease, cancer)               89 (10)         31 (41)         ***
Decrease saturated fat intake in
my diet                              89 (9)          64 (22)         ***
Live longer                          73 (25)         15 (49)         ***
Eat a greater variety of
interesting foods                    88 (8)          22 (33)         ***
Have plenty of energy                80 (19)         16 (47)         ***
Have a better quality of life        88 (11)         12 (38)         ***
Stay healthy                         94 (6)          25 (38)         ***
Be less aggressive                   41 (37)          6 (36)         ***
Be more content with myself          85 (12)          8 (36)         ***
Save money                           52 (24)         19 (37)         ***
Have a tastier diet                  90 (8)           7 (34)         ***
Satisfy my religious and/or
spiritual needs                      69 (11)          3 (19)         ***
Be fit                               72 (21)         17 (33)         ***
Eat more fruit and vegetables        94 (2)          73 (11)         ***
Be healthier by decreasing my
intake of chemicals, steroids
and antibiotics which are found
in meat                              93 (6)          27 (44)         ***
Increase my control over my own
health                               89 (8)          19 (39)         ***
Lower my chances of getting food
poisoning                            76 (17)         16 (34)         ***
Help the environment                 97 (3)          18 (36)         ***
Decrease hunger in the Third
World                                78 (15)         11 (40)         ***
Help animal welfare/rights           99 (1)          33 (32)         ***
Increase the efficiency of food
production                           85 (13)         12 (44)         ***
Help the feminist cause              21 (53)          3 (27)         ***
Help create a more peaceful
world                                82 (14)          6 (24)         ***

(a) Items used in categorical principal component analysis.
***P < 0.001

Table 3. Percent of self-identified vegetarians and non-vegetarians in
agreement with statements of perceived barriers to vegetarianism

                                 Vegetarians  Non-vegetarians
Item (a)                         (n = 103)      (n = 546)         P
                                 Percent in agreement (% unsure)

My friends eat meat                  18 (3)         32 (13)       ***
My family eats meat                  20 (4)         43 (10)       ***
I like eating meat                    1 (3)         83 (9)        ***
I would (or do) feel
conspicuous among others              8 (7)          9 (20)       **
There is not enough iron in
vegetarian diets                      4 (7)         29 (38)       ***
There is not enough protein in
vegetarian diets                      1 (4)         23 (38)       ***
I would be (or am) worried
about my health (other than
lack of iron or protein)              3 (5)         30 (29)       ***
Someone else decides on most of
the food I eat                        1 (2)         10 (7)        ***
My family/spouse/partner won't
eat vegetarian food                   5 (5)         31 (14)       ***
It takes too long to prepare
vegetarian food                       1 (0)          8 (34)       ***
I don't want to eat strange or
unusual foods                         0 (3)         22 (14)       ***
Vegetarian options are not
available where I shop or in
the canteen or at my home            12 (7)         13 (21)       **
I don't have enough willpower         2 (1)         20 (16)       ***
I think humans are meant to
eat meat                              0 (3)         47 (25)       ***
It is inconvenient                    5 (3)         23 (26)       ***
I don't know what to eat
instead of meat                       4 (1)         22 (17)       ***
I lack the right cooking skills       1 (0)         16 (15)       ***
I do not want to change my
eating habit or routine               4 (3)         59 (17)       ***
There is too limited a choice
when I eat out                       31 (6)         35 (22)       ***
People would (or do) think that
I'm a wimp or not 'macho'
enough                                5 (5)          4 (11)       NS (b)
Vegetarian diets are not
filling enough                        3 (1)         19 (31)       ***
Vegetarian diets are boring           1 (1)         24 (31)       ***
I don't want people to
stereotype me negatively (e.g.
that I must be strange)               5 (4)          8 (14)       **
I wouldn't (or don't) get
enough energy or strength from
the food                              2 (3)         17 (34)       ***
I need more information about
vegetarian diets                     14 (9)         42 (17)       ***

(a) Items used in categorical principal component analysis.
(b) NS not significant.
**P < 0.01; ***P < 0.001

Table 4. Percent of vegetarians, cognitive vegetarians and omnivores in
agreement with statements on health beliefs and sources of nutrition
information

                                                 Cognitive
                                   Vegetarians  vegetarians
      Item (a)                     (n = 103)      (n = 81)

Health beliefs and behaviour       % in agreement (% unsure)
I frequently look for information
on healthy eating.                 77 (11)      74 (9)
I do not need to make any changes
to the food I eat as it is
already healthy enough.            58 (16)      40 (12)

Health, food or nutrition
information sources
Use                                % sometimes or often use
Use dietitians/nutritionists       28           45
Use doctors                        23           49
Use health food shops/staff        45           40
Use vegetarian magazines           78           28
Use alternative health
practitioners                      44           29
Use cooking magazines              43           69

Trust                              % trust (% unsure)
Trust television programs          12 (32)      31 (35)
Trust newspaper articles           15 (43)      28 (44)
Trust advertising                   4 (12)      18 (31)
Trust dietitians/nutritionists     49 (41)      73 (19)
Trust doctors                      34 (27)      82 (13)
Trust National Heart
F'dn/Anti-Cancer Council           50 (31)      80 (15)
Trust health food shops/staff      50 (41)      46 (36)
Trust vegetarian magazines         81 (16)      44 (45)

Personal values                    % very important (% fairly important)
Protecting the environment
(preserving nature)                87 (13)      65 (31)
A world of beauty (beauty of
nature and the arts)               71 (26)      48 (45)
Helpful (working for the welfare
of others)                         44 (52)      55 (41)
Equality (equal opportunity for
all)                               87 (13)      65 (29)

Food supply issues                 % concerned (% unsure)
The links between food and cancer  86 (12)      80 (14)
Chemical additives in food         96 (4)       72 (8)

                                   Omnivores  P (b)   P (c)
      Item (a)                     (n = 468)

Health beliefs and behaviour       % in agreement (% unsure)
I frequently look for information
on healthy eating.                 58 (16)    NS (d)  *
I do not need to make any changes
to the food I eat as it is
already healthy enough.            45 (20)    **      NS

Health, food or nutrition
information sources
Use                                % sometimes or often use
Use dietitians/nutritionists       30         *       **
Use doctors                        46         ***     NS
Use health food shops/staff        24         NS      **
Use vegetarian magazines           11         ***     ***
Use alternative health
practitioners                      19         *       *
Use cooking magazines              56         ***     *

Trust                              % trust (% unsure)
Trust television programs          35 (44)    **      *
Trust newspaper articles           31 (53)    *       *
Trust advertising                  20 (43)    ***     NS
Trust dietitians/nutritionists     71 (27)    **      *
Trust doctors                      83 (15)    ***     NS
Trust National Heart
F'dn/Anti-Cancer Council           84 (15)    ***     **
Trust health food shops/staff      30 (54)    NS      **
Trust vegetarian magazines         23 (64)    ***     ***

Personal values                    % very important (% fairly important)
Protecting the environment
(preserving nature)                47 (47)    ***     *
A world of beauty (beauty of
nature and the arts)               23 (57)    **      ***
Helpful (working for the welfare
of others)                         33 (61)    NS      ***
Equality (equal opportunity for
all)                               60 (33)    ***     NS

Food supply issues                 % concerned (% unsure)
The links between food and cancer  68 (27)    NS      *
Chemical additives in food         79 (11)    ***     *

(a) Only a selection of the survey items is shown, in order to highlight
the characteristics of cognitive vegetarians.
(b) Comparisons between vegetarians and cognitive vegetarians.
(c) Comparisons between cognitive vegetarians and omnivores.
(d) NS = not significant.
* P < 0.05, **P < 0.01, *** P < 0.001

Acknowledgment 

This research was conducted in partial fulfilment of Emma Lea's PhD degree at the University of Adelaide, under the supervision of Anthony Worsley. Financial assistance for the survey was provided through an ARC small grant.

References

1. Vegetarian Resource Group. How many vegetarians are there? A 2000 National Zogby poll. Available from: URL: http://www.vrg.org. Accessed 31 March 2000.

2. Australian Bureau of Statistics. National Nutrition Survey Selected Highlights Australia 1995. Canberra: Australian Government Publishing Service; 1997.

3. Vegetarian Society UK. Summary of Realeat polls 1984-1999. Available from: URL: http://www.vegsoc.org/info/realeat.html. Accessed 9 April 2001.

4. American Dietetic Association: How many vegetarians are there? J Am Diet Assoc 1997;97:1287.

5. Gallup. The Realeat Survey 1997--changing attitudes to meat consumption. Newport Pagnell: Haldane Foods; 1997.

6. Baghurst K. Red meat consumption in Australia: intakes, contributions to nutrient intake and associated dietary patterns. Eur J Cancer Prev 1999;8:185-91.

7. Rosegrant MW, Leach N, Gerpacio RV. Alternative futures for world cereal and meat consumption. Proc Nutr Soc 1999;58:219-34.

8. Vegetarian Resource Group. How many people order vegetarian meals when eating out? A 1999 Zogby International poll. Available from: URL: http://www.vrg.org. Accessed 3 April 2000.

9. American Dietetic Association. Position of the American Dietetic Association: vegetarian diets. J Am Diet Assoc 1997;97:1317-21.

10. Key TJ, Fraser GE, Thorogood M, Appleby PN, Beral V, Reeves G, et al. Mortality in vegetarians and nonvegetarians: detailed findings from a collaborative analysis of 5 prospective studies. Am J Clin Nutr 1999;70(Suppl.):516S-24S.

11. US Department of Health and Human Services. Healthy People 2010--Conference Edition. Washington, DC: US Department of Health and Human Services; 2000.

12. World Cancer Research Fund, American Institute for Cancer Research. Food. Nutrition and the Prevention of Cancer: a global perspective. Washington, DC: American Institute for Cancer Research; 1997.

13. Appleby PN, Thorogood M, Mann JI, Key TJ. Low body mass index in non-meat eaters: the possible roles of animal fat, dietary fibre and alcohol. Int J Obes Relat Metab Disord 1998;22:454-60.

14. Li D, Sinclair A, Mann N, Turner A, Ball M, Kelly F, et al. The association of diet and thrombotic risk factors in healthy male vegetarians and meat-eaters. Eur J Clin Nutr 1999;53:612-9.

15. Thomas HV, Davey GK, Key TJ. Oestradiol and sex hormone-binding globulin in premenopausal and post-menopausal meat-eaters, vegetarians and vegans. Br J Cancer 1999;80:1470-5.

16. Haddad EH, Sabate J, Whitten CG. Vegetarian food guide pyramid: a conceptual framework. Am J Clin Nutr 1999;70:615S-9S.

17. Sanders TAB. The nutritional adequacy of plant-based diets. Proc Nutr Soc 1999;58:265-9.

18. Mann JI. Optimizing the plant-based diet. Asia Pac J Clin Nutr 2000;9(Suppl):60S-4S.

19. Alexander D, Ball MJ, Mann J. Nutrient intake and haematological status of vegetarians and age-sex matched omnivores. Eur J Clin Nutr 1994;48:538-46.

20. Dwyer JT. Health aspects of vegetarian diets. Am J Clin Nutr 1988;48:712-38.

21. Dillman DA. Mail and electronic surveys: the tailored design method. New York: J. Wiley: 2000.

22. Lea E, Worsley A. Influences on meat consumption in Australia. Appetite 2001;36:127-36.

23. Fiddes N. Meat: a natural symbol. London: Routledge; 1991.

24. Freeland-Graves J, Greninger SA, Graves GR, Young RK. Health practices, attitudes, and beliefs of vegetarians and nonvegetarians. J Am Diet Assoc 1986;86:913-8.

25. Richardson, NJ. UK consumer perceptions of meat. Proc Nutr Soc 1994;53:281-7.

26. Schwartz SH. Universals in the content and structure of values: theoretical advances and empirical tests in 20 countries. Adv Exp Soc Psychol 1992;25:1-65.

27. Worsley A. Baghurst K, Skrzypiec G. Meat Consumption and Young People. Adelaide: CSIRO Final Report to the Meat Research Corporation; 1995.

28. Kearney M, Gibney MJ, Martinez JA, de Almeida MDV, Friebe D, Zunft HJF, et al. Perceived need to alter eating habits among representative samples of adults from all member states of the European Union. Eur J Clin Nutr 1997;51(Suppl 2):30S-5S.

29. Worsley A, Scott V. Consumers' concerns about food and health in Australia and New Zealand. Asia Pac J Clin Nutr 2000;9:24-32.

30. Lea E, Worsley A. The cognitive contexts of beliefs about the healthiness of meat. Public Health Nutr 2002;5:37-45.

31. Rokeach M. The nature of human values. New York: Free Press; 1973.

32. Australian Bureau of Statistics. Basic Community Profile software. 1996 Census of Population and Housing. Canberra: Australian Bureau of Statistics; 1996.

33. Australian Bureau of Statistics. National Nutrition Survey 1995. Canberra: Australian Bureau of Statistics; 1995.

34. Richardson NJ, Shepherd R, Elliman N, Meat consumption, definition of meat and trust in information sources in the UK population and members of The Vegetarian Society. Ecol Food Nutr 1994;33:1-13.

35. Cashel KM, Crawford D, Deakin V. Milk choices made by women: what influences them, and does it impact on calcium intake? Public Health Nutr 2000;3:403-10.

36. Sapp SG, Harrod WJ. Social acceptability and intentions to eat beef: an expansion of the Fishbein-Ajzen model using reference group theory. Rural Sociol 1989;54:420-38.

37. Zey M, McIntosh WA. Predicting intent to consume beef: normative versus attitudinal influences. Rural Sociol 1992;57:250-65.

38. Furnham A, Forey J. The attitudes, behaviors and beliefs of patients of conventional vs. complementary (alternative) medicine. J Clin Psychol 1994;50:458-69.

39. Allen MW, Wilson M, Ng SH, Dunne M. Values and beliefs of vegetarians and omnivores. J Soc Psychol 2000;140:405-22.

40. Holm L. Mohl M. The role of meat in everyday food culture; an analysis of an interview study in Copenhagen. Appetite 2000;34:277-83.

41. Worsley A, Skrzypiec G. Teenagers' social attitudes and red meat consumption. In; Royal Society of New Zealand, editor. Proceedings of the Royal Society of New Zealand Miscellaneous Series 36; Wellington: Royal Society of New Zealand; 1996. p. 10-15.

42. Adams CJ. The sexual politics of meat: a feminist-vegetarian critical theory. New York: Continuum; 1990.

43. Fagerli RA, Wandel M. Gender differences in opinions and practices with regard to a 'healthy diet'. Appetite 1999;32:171-90.

44. Rappoport L, Peters G, Downey R. McCann T, Huff-Corzine L. Gender and age differences in food cognition. Appetite 1993;20:33-52.

45. Santich B. A brief history of food choice in Australia. In: Worsley A, editor. Multidisciplinary Approaches to Food Choice: Food Choice Conference, 1996; Adelaide: Department of Community Medicine, University of Adelaide; 1996. p. 1-3.

46. Mintz SW. Sweetness and power: the place of sugar in modern history. New York: Penguin Books; 1985.

School of Exercise and Nutrition Sciences, Deakin University, Victoria

E. Lea. BA (Hons), PhD, Postdoctoral Research Fellow

A. Worsley. BSc (Hons), PhD, Professor of Population Health

Correspondence: E. Lea, School of Exercise and Nutrition Sciences. Deakin University, 221 Burwood Highway, Burwood, Victoria 3125. Email: [e-mail]

E. Lea designed and implemented the survey, carried out the analysis and drafted the paper. A. Worsley contributed to conception of the survey, statistical methodology and data interpretation, as well as critical revision of the paper.